{"id":"W4285817857","doi":"10.1093/gigascience/giac065","title":"Toward global integration of biodiversity big data: a harmonized metabarcode data generation module for terrestrial arthropods","year":2022,"lang":"en","type":"article","venue":"GigaScience","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Horizon 2020 Framework Programme; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Fundación Bancaria Caixa d'Estalvis i Pensions de Barcelona; European Commission","keywords":"Big data; Biodiversity; Computer science; Data science; Data integration; Ecology; Data mining; Biology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04750742,0.001942323,0.001437076,0.00819983,0.002115805,0.007410246,0.005685694,0.002097599,0.004131252],"category_scores_gemma":[0.0671023,0.001759575,0.003411796,0.005348761,0.001569429,0.007743847,0.01240052,0.004618455,0.005281246],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001573132,"about_ca_system_score_gemma":0.008636283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003627196,"about_ca_topic_score_gemma":0.005500579,"domain_scores_codex":[0.9894165,0.003833193,0.001693055,0.002203713,0.002318296,0.0005352061],"domain_scores_gemma":[0.9554532,0.01040746,0.003724193,0.01328254,0.01430208,0.002830529],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0008431309,0.000606243,0.0437213,0.005755998,0.0012496,0.002085663,0.01223822,0.01986175,0.06175708,0.05686405,0.1685152,0.6265017],"study_design_scores_gemma":[0.0002997919,0.0005038325,0.02523903,0.003245269,0.0007766486,0.001432285,0.003135911,0.08749584,0.09815812,0.09588034,0.6830928,0.0007401129],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0075914,0.0004329091,0.9426565,0.001889774,0.0002528568,0.001276077,0.007753181,0.03555999,0.002587282],"genre_scores_gemma":[0.01159823,0.0002345521,0.963854,0.0005407787,0.00006367586,0.001105191,0.01752169,0.0042772,0.000804722],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04750742,"threshold_uncertainty_score":0.2512463,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2371889043425931,"score_gpt":0.2950674986009877,"score_spread":0.05787859425839459,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}